The Back Alley Archives

Investors have spent the last few weeks bemoaning the devastation to their portfolios caused by the stock market downturn, which if it does not produce recovery by year-end will have made 2008 the worst stock market year since 1937. Their misery would be compounded if they knew that next year, while it may avoid more than moderate stock market mayhem, is likely to produce the worst bond market carnage in US history.

By bond market carnage I am not referring to carnage in the market for securitized subprime mortgages, defaulted credit card receivables, Russian subordinated debt and Venezuelan trade paper. That has by and large already happened, although only a portion of the losses in those markets have already been admitted to – no more than $600 billion of the eventual total of perhaps $2.5-3 trillion in losses.

The rest of the market is taking Blackstone’s Steven Schwarzman’s approach of demanding that “market to market” accounting rules be reversed immediately. Tough, guys, you were happy enough to have the spurious mark-ups from “mark to market” in good years, which enabled you to pay yourselves fat bonuses without actually having earned anything. It’s only fair that the inflated prices at which your portfolios were valued at the top of the bubble should be marked down to reflect the new and unpleasant reality.

Next time, perhaps we can stick to the old rule that assets don’t get marked up in value until they are sold, but that clear impairment in value results in a markdown. It will mean fewer bonuses for Wall Street traders, but never mind, they’d only have to pay them all away in taxes – making Wall Streeters pay more tax was the principal “change” President-elect Obama and his supporters have been calling for. British experience in 1973-75, long before “mark to market” had been thought of was that eventually all excesses in financial institutions’ balance sheets must be paid for, and that once recession hits it’s quite possible to go bankrupt while presenting a balance sheet of unspotted solidity to the unsuspecting public.

The $2.5-3 trillion loss from value impairment of junk debt is however less than the value impairment that can be expected in the next year from the decrease in value of Treasury bonds and other prime quality debt. This prime debt has been used as a “safe haven” by investors for the last 20 years, and it is nothing of the kind.

Currently the 10-year Treasury bond yields a pathetic 3.78%, well towards the low end of its long-term historic range and well below the US inflation rate of about 5%. Including the housing finance agencies Fannie Mae and Freddie Mac debt of about $5 trillion (which is now explicitly government guaranteed) and the $4.3 trillion held in the social security trust fund, there is about $15.6 trillion of US federal debt, a total that increased by over $1 trillion in the year to September 2008 and is increasing even more rapidly currently. Add about $9 trillion of home mortgage debt and $6 trillion of high quality corporate debt (and ignoring debt of financial institutions, which can be expected to be largely matched against other debts) and you have a total outstanding amount of $30 trillion of debt subject to interest rate risk, excluding the junk and near-junk that is currently in the process of defaulting.

There are a number of reasons why Treasury bond yields and the yield curve in general are likely to rise sharply in 2009:

· Borrowing requirements. Treasury borrowed over $1 trillion in the year to September 2008; it is expected to borrow close to $2 trillion in the year to September 2009. That’s 13% of US Gross Domestic Product. Not all of this is deficit; about $500 billion is refinancing and another $500 billion is for bailout schemes, some of which the US taxpayer may eventually see back. Still, in terms of GDP that’s far more debt than the US capital market has ever been asked to absorb, other than during World War II. At some point, “crowding out” must occur; we certainly cannot assume that Asian central banks will want to take the entire load, at interest rates less than zero in real terms.
· Inflation. The Fed appears to believe that the current recession will bail the United States out of its inflation problem. The example is given of Japan in the late 1990s, after which the Fed explains that they will avoid the mistakes of the Bank of Japan, thus preventing damaging deflation. Actually that seems to be wrong on two counts. The main mistake in 1990s Japan was not monetary but fiscal; government spending was allowed to expand inexorably, producing ever larger and larger deficits. That mistake appears to be only too likely to be repeated here. The difference is that the United States currently has a 1% Federal Funds rate and 5% inflation, the approximate opposite of Japan in the early years of its slump. With M2 money supply (the one the Fed will divulge) up at an annual rate of 18.3% since the beginning of September it seems likely that inflation will accelerate – as it did in the recessions of 1973-74 and 1979-80.
· Rising real rates of return. The yields on Treasury Inflation Protected Securities have already risen from just over 1% to nearly 3% since the beginning of 2008. Given the excess of bonds coming to the market, it makes sense that real yields should rise. That in itself suggests that conventional Treasury bonds are hopelessly overvalued – with the 10-year TIPS yielding 2.82% and the 10 year Treasury 3.78%, the implied rate of US inflation over the decade to 2018 is 0.96% per annum, for a total rise in prices by 2018 of less than 10%. If you think that’s likely, I can get you a deal on Brooklyn Bridge!

Thus Treasury bond and other prime bond yields can be expected to rise sharply in 2009. This will cause losses to their holders. To the extent that such holders are foreign central banks, the United States probably doesn’t need to worry. Foreign central banks have been gentlemanly holders of US debt through periods such as 2002-08 when the dollar has depreciated; a rise in interest rates simply gives them another way of making a loss. Personally if I were the Chairman of the People’s Bank of China and Treasuries had lost me the kind of money they have in the last five years I’d probably declare war on the US, but fortunately central bankers are a phlegmatic and tolerant lot!

However domestic holders are a more serious problem. To the extent that pension funds have losses on their holdings of bonds, they will need to raise contributions; to the extent that insurance companies have such losses they will need to raise premiums. Some entities will be hedged, but by doing so they will have simply transferred the interest rate risk to somebody else; by definition of derivatives the total outstanding derivatives position must be zero, however large the individual positions taken.

Assuming the $30 trillion state, mortgage and private corporate debt outstanding has an average duration of 5 years, a fairly conservative assumption, and neither the shape of the yield curve nor the premiums payable for risk alter significantly by the end of 2009, a 1% rise to 4.74% in Treasury bond rates by December 2009 would cause a total loss to investors in the $30 trillion of Federal, agency, mortgage and prime corporate debt of 3.9% of the debt’s principal amount, or $1.17 trillion. Not as bad as the credit losses.

However once rates start rising, they are likely to rise much more than 1%. To cause a loss of $3 trillion, the same as the estimated credit losses, 10 year Treasury bond yields would have to rise to 6.43%. Hardly an excessive assumption; 10-year Treasuries yielded 6.44% on average during 1996, at the beginning of the Fed’s money bubble, in which year inflation was 3.4%.

More extreme moves are certainly possible. In 1990, 10-year Treasuries yielded an average of 8.55%, while inflation in that year was 6.3%. A rise in the yield curve to an 8.55% 10-year Treasury yield would cost investors $5.06 trillion, almost double the credit losses from subprime and its brethren. Should we revert fully to the days when Paul Volcker was Fed Chairman and get the 13.92% 10-year Treasury yield of 1981, a year in which inflation was 8.9%, the cost to investors from the interest rate rise alone (we can assume a few additional bankruptcies, I think) would by $9.33 trillion, about two thirds of the current value of common stocks outstanding and more than three times expected credit losses.

One can debate the probability of the various outcomes above. Inflation is already around 5% and is unlikely to drop much, so the 1996 estimate for the peak 10-year Treasury yield would seem low. On the other hand, while inflation could well reach 8.9%, it seems unlikely that we will need to push Treasury yields quite up to 1981’s Volckerian levels, at least not within the next year. So the 1990 estimate is perhaps the best, involving a loss to investors of around $5 trillion or a little over. Such a loss will produce fewer calls for bailouts than the $3 trillion credit losses, but just as much economic damage, albeit much of it unnoticed by the general public.

And it is still ahead of us!

Do you go looking for bad news AA?